💻 computer science

Privacy-preserving Blockchain-enabled Parametric Insurance via Remote Sensing and IoT

This paper proposes a privacy-preserving parametric insurance framework that leverages blockchain, IoT, and remote sensing data enhanced by optimized zero-knowledge proofs (zk-SNARKs) to automate claims verification while protecting user privacy and significantly reducing gas costs.

Original authors: Mingyu Hao, Keyang Qian, Sid Chi-Kin Chau

Published 2026-05-04

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Mingyu Hao, Keyang Qian, Sid Chi-Kin Chau

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you have a special type of insurance called Parametric Insurance. Unlike traditional insurance, where you have to wait for an adjuster to come to your house, take photos, argue about the damage, and fill out mountains of paperwork, parametric insurance is like a vending machine.

Here's how it works: You set a rule. For example, "If a bushfire burns more than 50% of my land, the insurance pays out automatically." The machine checks a public sensor (like a satellite) to see if the rule was met. If the sensor says "Yes," the money drops into your account instantly. No paperwork, no waiting.

The Problem: The Glass House
The paper explains that while this sounds great, putting this system on a Blockchain (a public, digital ledger) creates a new problem. Think of a blockchain as a giant glass house. Everyone can see inside.

Privacy Issue: If you claim insurance for a bushfire, everyone on the blockchain can see exactly where your house is and that you are claiming money. It's like shouting your address and bank balance to the whole neighborhood.
Cost Issue: Doing complex math on the blockchain is expensive. It's like paying a high "gas fee" for every calculation. Checking a huge satellite image to see if a fire happened is too heavy and costly for the blockchain to do directly.

The Solution: The Magic Envelope (zk-SNARKs)
The authors propose a solution using a cryptographic trick called zk-SNARKs (Zero-Knowledge Succinct Non-Interactive Arguments of Knowledge).

Imagine you want to prove to a bouncer that you are over 21 without showing your ID card or revealing your actual age. You hand the bouncer a Magic Envelope.

The Proof: Inside the envelope is a mathematical proof that says, "I am over 21."
The Secrecy: The bouncer can open the envelope and verify the math is correct, but they cannot see your ID, your name, or your exact age. They just know the statement is true.
The Speed: The envelope is tiny. It's much smaller than your actual ID card, so it's cheap and fast to check.

How the Paper's System Works
The authors built a system where:

The Insurer sets up the rules (the "vending machine") on the blockchain.
The Data Providers (satellite companies or IoT sensors) take a picture of the land. They sign the picture to prove it's real, but they don't tell the blockchain which specific land it is yet.
The Insured Person takes the satellite data and runs a complex calculation off-chain (outside the expensive blockchain) to see if the fire rule was met.
The Magic Envelope: The insured person creates a Zero-Knowledge Proof. This proof tells the blockchain: "I have a signed satellite picture from a trusted source, and the math proves a fire happened on the land I insured."
- The blockchain checks the math.
- The blockchain verifies the signature is real.
- Crucially: The blockchain never sees the actual location coordinates or the raw image data. Your privacy is safe.

The "Sonic" Upgrade
The paper uses a specific type of Magic Envelope technology called Sonic. The authors made two big improvements to Sonic:

Mixing Different Sources: Sometimes you need data from two different satellites or sensors. The authors figured out how to combine these different "trusted sources" into one Magic Envelope without them needing to know about each other. It's like having two different witnesses sign a document without them ever meeting.
The Bulk Discount (Batch Verification): Checking many proofs one by one is slow and expensive. The authors created a way to check multiple proofs at once, like a "bulk discount" for the blockchain. This cut the cost of verifying the claim by 80%.

The Real-World Test: Bushfires
To prove this works, the team built a prototype for Bushfire Insurance on the Ethereum blockchain (a real, public blockchain).

They used real satellite images of Australian bushfires.
They calculated the "burn severity" (how bad the fire was) using a formula called dNBR.
They successfully generated the "Magic Envelope" proof that a fire happened without revealing the specific farm's location.
They verified it on the blockchain for a tiny fraction of the usual cost (dropping the fee from about $39 to roughly $8).

In Summary
This paper presents a way to automate insurance payouts using satellites and smart contracts, but with a "Magic Envelope" that keeps your private data (like your location) hidden from the public. It also makes the process cheap enough to be practical by inventing a faster way to check the math.

Technical Summary: Privacy-preserving Blockchain-enabled Parametric Insurance via Remote Sensing and IoT

1. Problem Statement

Traditional indemnity insurance suffers from high operational costs, opacity, inefficiency, and a lack of trust due to manual case-by-case assessments. While parametric insurance—which automates payouts based on verifiable indices derived from public data (e.g., IoT, remote sensing)—offers a solution, integrating it with blockchain introduces two critical challenges:

Privacy Leakage: Blockchain's transparency mandates that transaction records and processing data be publicly visible. In parametric insurance, this exposes sensitive user data (e.g., specific insured locations or claim details) to the public ledger, potentially allowing deanonymization and compromising user privacy.
High Processing Costs: Smart contract execution on platforms like Ethereum is metered by "gas fees," which scale with computational complexity and memory usage. Complex tasks, such as processing large satellite imagery for bushfire detection, are prohibitively expensive to execute directly on-chain.

Existing solutions, such as pseudo-anonymity or permissioned blockchains, fail to provide strong privacy guarantees while maintaining the transparency and accountability required for decentralized insurance.

2. Methodology

The authors propose a privacy-preserving parametric insurance framework that leverages Succinct Zero-Knowledge Proofs (zk-SNARKs) to enable transparent verification without revealing private data. The core protocol allows an insuree to submit a zero-knowledge proof to a blockchain smart contract, attesting to the validity of a claim and the authenticity of its data sources without disclosing the underlying private inputs.

Core Protocol: Extended Sonic zk-SNARK

The framework is built upon the Sonic zk-SNARK protocol, which supports universal and updatable structured reference strings (SRS). The authors introduce two critical extensions to Sonic to address the specific needs of parametric insurance:

Authentication of Heterogeneous Data Sources:
- In real-world scenarios, data comes from multiple independent providers (e.g., different satellite imagery providers or IoT sensors) who are agnostic to the specific insurance application.
- The authors extend Sonic to allow a prover to incorporate data from $J$ independent sources, each with its own Structured Reference String (SRS) and digital signature.
- The protocol verifies that the data used in the claim computation is authentic (signed by the provider) and correctly incorporated into the proof, without requiring the data providers to coordinate with the circuit designer (the insurer). This prevents collusion and ensures data impartiality.
Efficient On-chain Processing via Batch Verification:
- Standard Sonic protocols require verifying multiple polynomial openings at different points, which incurs high gas costs due to multiple pairing checks.
- The authors propose an enhanced polynomial commitment scheme that enables batch verification. This allows the verifier to check multiple polynomial openings simultaneously with a single proof, significantly reducing the number of pairing equations required on-chain.

Application: Bushfire Parametric Insurance

The framework is instantiated for bushfire insurance, where the claim validity depends on the delta Normalized Burn Ratio (dNBR) calculated from satellite imagery.

Data Processing: The system calculates dNBR using Near-Infrared (NIR) and Shortwave Infrared (SWIR) bands from satellite images (e.g., Sentinel-2).
Constraint System: The logic for determining "severe burn" (e.g., pixels with dNBR > threshold) is encoded into a constraint system involving linear and multiplicative constraints.
Location Privacy: To protect the insured location, the data provider signs a hash of the location combined with the data commitment. The smart contract verifies the signature against the hash, ensuring the data corresponds to the correct location without revealing the location itself on-chain.

3. Key Contributions

Extended zk-SNARK Protocol (Sdat): An extension of the Sonic protocol that supports the validation of data from heterogeneous and independent sources. This allows data providers to authenticate data independently of the application logic, preventing collusion between data sources and insurers.
Batch Verification Scheme: A novel polynomial commitment scheme designed for Sonic that enables batch verification of multiple polynomial openings. This reduces the on-chain verification gas cost by up to 80% compared to standard implementations.
Generic Framework: A proposed generic framework for privacy-preserving parametric insurance that can be applied to various scenarios (e.g., flight delay, flooding) beyond bushfires.
Proof-of-Concept Implementation: A working prototype implemented on the Ethereum blockchain (Goerli testnet) for bushfire insurance, utilizing Solidity smart contracts for on-chain verification.

4. Experimental Results

The authors conducted extensive empirical evaluations on the Ethereum testnet using satellite imagery data from the Digital Earth Australia (DEA) repository.

Scalability: The proving time scales linearly with the input size (number of pixels), while the proof size remains constant (approx. 1.22 KB). Verification time on-chain is also constant regardless of input size.
Gas Cost Reduction:
- The Enhanced Verifier (using batch verification) reduced verification gas costs by approximately 78% compared to the standard Sonic verifier.
- The Enhanced+ Verifier (which stores SRS elements off-chain and retrieves them via oracles) achieved an 80% reduction in gas costs.
- Specifically, the verification cost for a claim dropped from ~1.62M gas (Standard Sonic) to ~0.34M gas (Enhanced), reducing the USD cost from ~$38.91 to ~$8.19 (based on 2022 rates).
Overhead: The Structured Reference String (SRS) size is significant (~191 MB) but is generated once and shared. Only a fixed number of SRS elements are required for on-chain verification.

5. Significance and Claims

The paper claims to offer a practical solution to the trade-off between privacy and transparency in blockchain-based insurance. By utilizing zk-SNARKs, the framework:

Protects User Privacy: Ensures that sensitive claim data (location, specific sensor readings) remains hidden from the public ledger while still allowing for public verification of the claim's validity.
Ensures Data Integrity: Validates the authenticity of data from multiple independent sources, preventing fraud and ensuring the impartiality of the data used for payouts.
Improves Economic Viability: The proposed optimizations (batch verification and off-chain SRS management) drastically reduce the computational costs (gas fees) associated with on-chain verification, making complex parametric insurance products economically feasible on public blockchains like Ethereum.

The authors conclude that their framework demonstrates the feasibility of deploying privacy-preserving, automated insurance systems on real-world blockchain platforms, paving the way for broader adoption in financial risk management.